City Research Online

Abstract

In the current digital era, supply chain companies possess vast amounts of data that can drive smarter products and services. Machine Learning (ML) and Artificial Intelligence (AI) can extract valuable insights from this data. Model buyers often encounter high costs or uncertainty about the quality of models developed by large corporations, while many small ML developers have developed suitable models. These small owners are often reluctant to share or compete due to concerns about transparency and fairness in the evaluation process. Decentralised model marketplaces aim to bridge this gap, addressing issues related to cost, privacy, and fairness, but significant barriers still hinder participation. This paper introduces a decentralised evaluation platform of the ML model marketplace that tackles these marketplace challenges. The transparent, fair, and immutable ranking system enables buyers to find models that meet their budget and quality expectations and encourages model owners to share their best models. The proposed workflow is implemented and analysed for efficiency, throughput, and costs.

Publication Type:	Conference or Workshop Item (Paper)
Additional Information:	© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Publisher Keywords:	Model Marketplace, Blockchain, Smart Contract, Machine Learning Model Evaluation, Supply Chain Data
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments:	School of Science & Technology School of Science & Technology > Computer Science School of Science & Technology > Engineering
SWORD Depositor:	Symplectic Administrator

Export

Downloads